Aims and scope

CCF Transactions on High Performance Computing (CCF THPC) is dedicated to publishing rigorously peer-reviewed, high-quality articles covering all aspects of supercomputing and high performance computing (HPC). Bringing together insights on, innovations in and implementations of HPC, the journal serves as an academic publication and communication platform and welcomes contributions from researchers, professionals, and industrial practitioners. The journal publishes original research and review articles on all aspects of high performance computing from design, service, and theory, to its practical use and applications. In addition, it addresses interdisciplinary topics like operating systems, networking and software engineering.

This journal is published on behalf of the China Computer Federation (CCF) and is supported by the CCF technical committee on High Performance Computing, and CCF technical committee on Computer Architectures and Information Storage Technology.

Specific topics of interest include, but are not limited to:

1) Architectures and related processors

Processor architecture (multicore, GPU, custom logic),
Novel computer architecture,
Reconfigurable architecture,
Special purpose architecture,
Processor micro-architecture,
Memory hierarchy,
Cache coherence and memory consistency model

2) Supercomputing systems

Hardware/software co-design,
Communication middleware,
Resilience and reliability technologies,
Interconnection,
Languages and compilers,
Power-efficient, energy-efficient systems/tools,
Performance and energy efficiency/benchmarks,
System cooling,
Large-scale system management,
Scalable and broadly applicable high performance algorithm/optimization

3) Storage

Parallel I/O,
High-performance file systems,
Solid state storage technologies and systems (e.g., flash, byte-addressable NVM),
Parallel storage system,
Memory-only storage systems,
Reliability, availability, and fault tolerance,
I/O performance tuning

4) Related applications of HPC

Parallel algorithms, models, frameworks Implementation,
HPC applications in atmospheric, materials, astrophysics, biology, AIs, and so on

Special Issue on High Performance Computing for Large Models
Submission Date: 2024-07-31

Recent advances in deep learning are driven by the growing scale of models and data. However, training and inference for large models, such as LLM, require massive computing power and incur high cost. In this special issue, we focus on the research of high-performance computing for large-scale models, such as parallel algorithms, frameworks, tools, systems, and co-design techniques for compute-efficient large models. We welcome paper submissions on all topics related to high-performance computing for large models. Topics of interest The special issue is dedicated to high-performance computing for large models. Topics of interest include, but are not limited to, the following areas about large models: • Hardware Efficient Operators • Parallel Strategies • Automatic Distributed Deep Learning • Sparsification and Quantization • Fault Tolerance for Large Models • Memory Saving Techniques • Performance Modelling • Deep Learning Compilers • Interconnected Networks • Communication Optimizations • Accelerators for AI

Special Issue on Intelligent Computing Systems
Submission Date: 2024-07-31

With the rapid advancement of computing technologies, an increasing and significant amount of research has focused on the efficiency and performance optimization of intelligent computing systems. In this special issue, we aim to cover challenges that can impact the efficiency and application performance of intelligent computing systems, especially in areas such as intelligent accelerator design, high performance computing, design automation for accelerator.etc. We welcome paper submissions on all topics related to intelligent computing systems and algorithm optimization. Topics of interest The special issue is dedicated to parallel system and algorithm optimization. Topics of interest include, but are not limited to: • •New hardware designs for high-performance intelligent computing • •Systems software and programming framework for intelligent computing • •Parallel and distributed computing for intelligent services • •Highly efficient machine learning and big data platforms • •New communication paradigms for intelligent computing systems • •Embedded and pervasive systems and high-performance AIoT • •Resource management and scheduling for intelligent applications • •Performance modeling and benchmarking for intelligent computing • •Optimization for power, energy, reliability, availability, and scalability • •Novel post-Moore, unconventional computing technologies

Special Issue on Convergence of Supercomputing and Intelligent Computing / Supercomputer Internet
Submission Date: 2024-08-10

The evolution of computing infrastructures catalyzes the swift progress of both technological innovation and industrial expansion. A burgeoning need for intelligence fuels the convergence of computing and network services, encompassing supercomputing and intelligent computing. In this special issue, we aim to explore the forefront of computility scheduling technologies and applications within the convergence of supercomputing and intelligent computing. This includes supercomputer Internet, intelligent computing systems, distributed machine learning algorithms, and large-scale supercomputing applications. We welcome paper submissions on all topics related to computing and network convergence, especially the fusion of diversified HPC. Topics of interest The special issue is dedicated to the convergence of supercomputing and intelligent computing. Topics of interest include, but are not limited to: • Computing and Network Convergence • Convergence of Supercomputing and Intelligent Computing • Computility Network and Supercomputer Internet • Intelligent Computing System • Large-scale Supercomputing Applications • Distributed Machine Learning: System and Algorithm

Special Issue on Scalable Performance Analysis and Optimization
Submission Date: 2024-08-15

As systems grow in complexity and scale, efficiently analyzing and optimizing application performance becomes crucial for sustaining the accelerated innovation on both scientific and AI research. In this special issue, we focus on the research on measuring, analyzing, diagnosing, optimizing the performance issues encountered by diverse applications running on large scale systems, from supercomputers to AI systems, to ensure high performance and energy efficiency . We welcome paper submissions on all topics related to scalable performance analysis and optimization. Topics of interest The special issue is dedicated to scalable performance analysis and optimization. Topics of interest include, but are not limited to: • Tools and techniques for performance analysis and diagnosis • Scalable performance modeling • High-performance computing (HPC) optimization • Energy-efficient performance optimization • Performance issues in big data and AI systems • Efficient performance auto-tuning techniques

Special Issue on Technology Advances and Applications of High-Performance and Reliable Storage Systems
Submission Date: 2024-09-10

High-performance and reliable storage systems have undergone significant advances in recent years, propelled by revolutionary technological breakthroughs which have expanded our capacity to generate and process large volumes of data. This rapid evolution has spurred groundbreaking progress across various fields, spanning from data-intensive applications to the expansive realms of AI and cloud computing, thus pushing the boundaries of scientific research, industrial applications, and everyday technology. In recognition of these advancements, we are delighted to unveil a special issue dedicated to showcasing the latest technology developments and applications of high-performance and reliable storage systems. This special issue focuses on examining various solutions and approaches for addressing the challenges in designing, implementing, and applying high-performance and reliable storage systems, such as emerging storage technologies, data management techniques, storage security and privacy, performance evaluation and tools, which can improve the efficiency and reliability of next-generation storage systems.